Recent genetic studies identified a variant in TREM2 (R47H) that doubles the risk for Alzheimer's disease (AD), similar to that of APOE?4, the strongest known genetic risk factor for late-onset AD. We performed deep resequencing of TREM2 in a large European-American (EA) and African-American (AA) population and found an enrichment of TREM2 coding variants in AD cases compared to controls, indicating that additional TREM2 variants affect AD risk. Additionally, recent studies in small datasets found that soluble TREM2 (sTREM2) levels in the cerebrospinal (CSF) are increased in AD cases compared to controls. Furthermore, CSF sTREM2 levels strongly correlate with CSF tau levels. These studies suggest that CSF sTREM2 levels may reflect biological events that link amyloid deposition and neurofibrillary tangle formation to cognitive decline. Studies from our group indicate that CSF sTREM2 levels are an informative phenotype for genetic studies. While these preliminary studies are promising, based on past successes of this approach, larger datasets will provide us with the power to identify novel genetic modifiers of CSF sTREM2. In this proposal, we will analyze a very large (n=3,476) and well-characterized dataset with extensive pre-existing CSF biomarker (A?, Tau, ptau and sTREM2), clinical and genetic data. The aims of the project are: 1) to identify single (common and rare) variants, genes and pathways associated with CSF sTREM2 levels; 2) to determine the role of those variants and genes on AD risk, onset and progression and 3) to perform functional analyses in iPSC-derived human microglia to determine the mechanisms by which the genetic variants identified in the genetic analyses affect sTREM2 levels and microglia function.